Rational inattention and tonic dopamine

Bayesian Reinforcement Learning With Limited Cognitive Load

All biological and artificial agents must act given limits on their ability to acquire and process information. As such, a general theory of adaptive behavior should be able to account for the complex interactions between an agent's learning history, decisions, and capacity constraints. Recent work in computer science has begun to clarify the principles that shape these dynamics by bridging ideas from reinforcement learning, Bayesian decision-making, and rate-distortion theory. This body of work provides an account of capacity-limited Bayesian reinforcement learning, a unifying normative framework for modeling the effect of processing constraints on learning and action selection. Here, we provide an accessible review of recent algorithms and theoretical results in this setting, paying special attention to how these ideas can be applied to studying questions in the cognitive and behavioral sciences.

Striatal dopamine integrates cost, benefit, and motivation

Striatal dopamine (DA) release has long been linked to reward processing, but it remains controversial whether DA release reflects costs or benefits and how these signals vary with motivation. Here, we measure DA release in the nucleus accumbens (NAc) and dorsolateral striatum (DLS) while independently varying costs and benefits and apply behavioral economic principles to determine a mouse's level of motivation. We reveal that DA release in both structures incorporates both reward magnitude and sunk cost. Surprisingly, motivation was inversely correlated with reward-evoked DA release. Furthermore, optogenetically evoked DA release was also heavily dependent on sunk cost. Our results reconcile previous disparate findings by demonstrating that striatal DA release simultaneously encodes cost, benefit, and motivation but in distinct manners over different timescales. Future work will be necessary to determine whether the reduction in phasic DA release in highly motivated animals is due to changes in tonic DA levels.

Tonic dopamine and biases in value learning linked through a biologically inspired reinforcement learning model

A hallmark of various psychiatric disorders is biased future predictions. Here we examined the mechanisms for biased value learning using reinforcement learning models incorporating recent findings on synaptic plasticity and opponent circuit mechanisms in the basal ganglia. We show that variations in tonic dopamine can alter the balance between learning from positive and negative reward prediction errors, leading to biased value predictions. This bias arises from the sigmoidal shapes of the dose-occupancy curves and distinct affinities of D1- and D2-type dopamine receptors: changes in tonic dopamine differentially alters the slope of the dose-occupancy curves of these receptors, thus sensitivities, at baseline dopamine concentrations. We show that this mechanism can explain biased value learning in both mice and humans and may also contribute to symptoms observed in psychiatric disorders. Our model provides a foundation for understanding the basal ganglia circuit and underscores the significance of tonic dopamine in modulating learning processes.

Dopamine regulates decision thresholds in human reinforcement learning in males

Dopamine fundamentally contributes to reinforcement learning, but recent accounts also suggest a contribution to specific action selection mechanisms and the regulation of response vigour. Here, we examine dopaminergic mechanisms underlying human reinforcement learning and action selection via a combined pharmacological neuroimaging approach in male human volunteers (n = 31, within-subjects; Placebo, 150 mg of the dopamine precursor L-dopa, 2 mg of the D2 receptor antagonist Haloperidol). We found little credible evidence for previously reported beneficial effects of L-dopa vs. Haloperidol on learning from gains and altered neural prediction error signals, which may be partly due to differences experimental design and/or drug dosages. Reinforcement learning drift diffusion models account for learning-related changes in accuracy and response times, and reveal consistent decision threshold reductions under both drugs, in line with the idea that lower dosages of D2 receptor antagonists increase striatal DA release via an autoreceptor-mediated feedback mechanism. These results are in line with the idea that dopamine regulates decision thresholds during reinforcement learning, and may help to bridge action selection and response vigor accounts of dopamine.

Undermatching Is a Consequence of Policy Compression

The matching law describes the tendency of agents to match the ratio of choices allocated to the ratio of rewards received when choosing among multiple options (Herrnstein, 1961). Perfect matching, however, is infrequently observed. Instead, agents tend to undermatch or bias choices toward the poorer option. Overmatching, or the tendency to bias choices toward the richer option, is rarely observed. Despite the ubiquity of undermatching, it has received an inadequate normative justification. Here, we assume agents not only seek to maximize reward, but also seek to minimize cognitive cost, which we formalize as policy complexity (the mutual information between actions and states of the environment). Policy complexity measures the extent to which the policy of an agent is state dependent. Our theory states that capacity-constrained agents (i.e., agents that must compress their policies to reduce complexity) can only undermatch or perfectly match, but not overmatch, consistent with the empirical evidence. Moreover, using mouse behavioral data (male), we validate a novel prediction about which task conditions exaggerate undermatching. Finally, in patients with Parkinson's disease (male and female), we argue that a reduction in undermatching with higher dopamine levels is consistent with an increased policy complexity.SIGNIFICANCE STATEMENT The matching law describes the tendency of agents to match the ratio of choices allocated to different options to the ratio of reward received. For example, if option a yields twice as much reward as option b, matching states that agents will choose option a twice as much. However, agents typically undermatch: they choose the poorer option more frequently than expected. Here, we assume that agents seek to simultaneously maximize reward and minimize the complexity of their action policies. We show that this theory explains when and why undermatching occurs. Neurally, we show that policy complexity, and by extension undermatching, is controlled by tonic dopamine, consistent with other evidence that dopamine plays an important role in cognitive resource allocation.

Vigilance, arousal, and acetylcholine: Optimal control of attention in a simple detection task

Paying attention to particular aspects of the world or being more vigilant in general can be interpreted as forms of ‘internal’ action. Such arousal-related choices come with the benefit of increasing the quality and situational appropriateness of information acquisition and processing, but incur potentially expensive energetic and opportunity costs. One implementational route for these choices is widespread ascending neuromodulation, including by acetylcholine (ACh). The key computational question that elective attention poses for sensory processing is when it is worthwhile paying these costs, and this includes consideration of whether sufficient information has yet been collected to justify the higher signal-to-noise ratio afforded by greater attention and, particularly if a change in attentional state is more expensive than its maintenance, when states of heightened attention ought to persist. We offer a partially observable Markov decision-process treatment of optional attention in a detection task, and use it to provide a qualitative model of the results of studies using modern techniques to measure and manipulate ACh in rodents performing a similar task.

Paying attention to a stimulus is costly, both in terms of energy and the lost opportunity to pay attention to something else. It is also beneficial, providing more information about its target. Thus, whether and when we pay more or less attention may best be considered as a choice of internal action that responds to this trade-off. Furthermore, measurements and manipulation of the neuromodulator acetylcholine have suggested that it is one of the instruments of attention, providing us with a window onto this choice. Here, we build an abstract model of a task in which an animal must look out for a brief visual stimulus that may or may not occur on each trial. We show that optimal attentional choices in the model depend on many factors, including how likely a signal is to occur across time, the balance between the improvement in information possible by paying greater attention and its increased cost, and whether there are also costs associated with switching between different attentional states. We also show that our model can qualitatively match results from experiments involving acetylcholine.

Effects of average reward rate on vigor as a function of individual variation in striatal dopamine

RATIONALE

We constantly need to decide not only which actions to perform, but also how vigorously to perform them. In agreement with an earlier theoretical model, it has been shown that a significant portion of the variance in our action vigor can be explained by the average rate of rewards received for that action. Moreover, this invigorating effect of average reward rate was shown to vary with within-subject changes in dopamine, both in human individuals and experimental rodents.

OBJECTIVES

Here, we assessed whether individual differences in the effect of average reward rate on vigor are related to individual variation in a stable measure of striatal dopamine function in healthy, unmedicated participants.

METHODS

Forty-four participants performed a discrimination task to test the effect of average reward rate on response times to index vigor and completed an [¹⁸F]-DOPA PET scan to index striatal dopamine synthesis capacity.

RESULTS

We did not find an interaction between dopamine synthesis capacity and average reward rate across the entire group. However, a post hoc analysis revealed that participants with higher striatal dopamine synthesis capacity, particularly in the nucleus accumbens, exhibited a stronger invigorating effect of average reward rate among the 30 slowest participants.

CONCLUSIONS

Our findings provide converging evidence for a role of striatal dopamine in average reward rate signaling, thereby extending the current literature on the mechanistic link between average reward rate, vigor, and dopamine.

Dopamine firing plays a dual role in coding reward prediction errors and signaling motivation in a working memory task

Little is known about how dopamine (DA) neuron firing rates behave in cognitively demanding decision-making tasks. Here, we investigated midbrain DA activity in monkeys performing a discrimination task in which the animal had to use working memory (WM) to report which of two sequentially applied vibrotactile stimuli had the higher frequency. We found that perception was altered by an internal bias, likely generated by deterioration of the representation of the first frequency during the WM period. This bias greatly controlled the DA phasic response during the two stimulation periods, confirming that DA reward prediction errors reflected stimulus perception. In contrast, tonic dopamine activity during WM was not affected by the bias and did not encode the stored frequency. More interestingly, both delay-period activity and phasic responses before the second stimulus negatively correlated with reaction times of the animals after the trial start cue and thus represented motivated behavior on a trial-by-trial basis. During WM, this motivation signal underwent a ramp-like increase. At the same time, motivation positively correlated with accuracy, especially in difficult trials, probably by decreasing the effect of the bias. Overall, our results indicate that DA activity, in addition to encoding reward prediction errors, could at the same time be involved in motivation and WM. In particular, the ramping activity during the delay period suggests a possible DA role in stabilizing sustained cortical activity, hypothetically by increasing the gain communicated to prefrontal neurons in a motivation-dependent way.

Impulsivity and risk-seeking as Bayesian inference under dopaminergic control

Bayesian models successfully account for several of dopamine (DA)'s effects on contextual calibration in interval timing and reward estimation. In these models, tonic levels of DA control the precision of stimulus encoding, which is weighed against contextual information when making decisions. When DA levels are high, the animal relies more heavily on the (highly precise) stimulus encoding, whereas when DA levels are low, the context affects decisions more strongly. Here, we extend this idea to intertemporal choice and probability discounting tasks. In intertemporal choice tasks, agents must choose between a small reward delivered soon and a large reward delivered later, whereas in probability discounting tasks, agents must choose between a small reward that is always delivered and a large reward that may be omitted with some probability. Beginning with the principle that animals will seek to maximize their reward rates, we show that the Bayesian model predicts a number of curious empirical findings in both tasks. First, the model predicts that higher DA levels should normally promote selection of the larger/later option, which is often taken to imply that DA decreases 'impulsivity,' and promote selection of the large/risky option, often taken to imply that DA increases 'risk-seeking.' However, if the temporal precision is sufficiently decreased, higher DA levels should have the opposite effect-promoting selection of the smaller/sooner option (higher impulsivity) and the small/safe option (lower risk-seeking). Second, high enough levels of DA can result in preference reversals. Third, selectively decreasing the temporal precision, without manipulating DA, should promote selection of the larger/later and large/risky options. Fourth, when a different post-reward delay is associated with each option, animals will not learn the option-delay contingencies, but this learning can be salvaged when the post-reward delays are made more salient. Finally, the Bayesian model predicts correlations among behavioral phenotypes: Animals that are better timers will also appear less impulsive.

Striatal dopamine synthesis capacity reflects smartphone social activity

Striatal dopamine and smartphone behavior have both been linked with behavioral variability. Here, we leverage day-to-day logs of natural, unconstrained smartphone behavior and establish a correlation between a measure of smartphone social activity previously linked with behavioral variability and a measure of striatal dopamine synthesis capacity using [¹⁸F]-DOPA PET in (N = 22) healthy adult humans. Specifically, we find that a higher proportion of social app interactions correlates with lower dopamine synthesis capacity in the bilateral putamen. Permutation tests and penalized regressions provide evidence that this link between dopamine synthesis capacity and social versus non-social smartphone interactions is specific. These observations provide a key empirical grounding for current speculations about dopamine's role in digital social behavior.

•

Putamen dopamine synthesis capacity correlates with smartphone social app use.

•

The correlation parallels a prior link between social app use and motor variability.

•

It is selective to social app use, controlling for multiple smartphone use factors.